Community Dent Oral Epidemiol 2014; 42; 289–299 All rights reserved
Ó 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Factors associated with dental caries in Brazilian children: a multilevel approach
Milene T. Martins1, Fernanda Sardenberg1, Mauro H. Abreu2, Miriam P. Vale1, Saul M. Paiva1 and Isabela A . Pordeus1 1 Department of Pediatric Dentistry and Orthodontics, Faculty of Dentistry, Federal University of Minas Gerais, Belo Horizonte, Brazil, 2Department of Social and Preventive Dentistry, Faculty of Dentistry, Federal University of Minas Gerais, Belo Horizonte, Brazil
Martins MT, Sardenberg F, Abreu MH, Vale MP, Paiva SM, Pordeus IA. Factors associated with dental caries in Brazilian children: a multilevel approach. Community Dent Oral Epidemiol 2014; 42: 289–299. © 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: The aims of the present study were to assess caries experience among Brazilian children aged eight to 10 years and determine social factors of this disease, through a multilevel approach. Methods: This is a cross-sectional study in which a representative sample of 1204 eight-to-tenyear-old children were randomly selected from 19 public and private schools in the city of Belo Horizonte, Brazil. After formal authorization from the parents, the children were clinically examined at school by two calibrated examiners (Kappa=0.78). Individual socioeconomic status was determined using household income, number of residents in the home and parents’/caregivers’ schooling. The social vulnerability index and type of school were used as contextual variables. The outcome variables were severity score of dental caries in permanent teeth (DMFT index), dental caries experience in permanent teeth (DMFT = 0 or 6¼ 0), severity score of dental caries in primary teeth (dmft index) and dental caries experience in primary teeth (dmft = 0 or 6¼ 0). Results: The minority of children (37.2%) had dental caries experience. Chi-squared and Mann–Whitney tests showed that age, parents’/caregivers’ schooling, household income and type of school were associated with dental caries experience in permanent and primary teeth (P < 0.05). Mann–Whitney, Kruskal–Wallis tests and Spearman correlation coefficient showed that age, parents’/caregivers’ schooling, household income, and type of school were associated with severity score of dental caries in permanent and primary teeth (P < 0.05). In multilevel approach, the individual variables, age and household income demonstrated association with dental caries experience and with severity score of dental caries in primary teeth (P < 0.05). Type of school was also associated with dental caries experience and with severity score of dental caries in these teeth (P < 0.05).In the Poisson multilevel approach, the analysis of the final estimation of variance components of the null model revealed a significant difference among the 19 schools with regard to severity score of dental caries in permanent teeth too (P < 0.001). Younger children and attending private schools showed lower severity score of dental caries in permanent teeth. Conclusion: Dental caries experience and severity score of dental caries in primary teeth was influenced by age, household income, and the type of school The dental caries experience in permanent teeth was influenced by age, parents’/caregivers’ schooling, household income, and type of school, whereas the severity score of dental caries in permanent teeth was influenced by age and type of school.
Dental caries is highly prevalent in children (1, 2) and one of the most prevalent chronic diseases worldwide (3). Despite the reported decline of dendoi: 10.1111/cdoe.12087
Key words: dental caries; multilevel approach; social determinants MT Martins, Department of Pediatric Dentistry and Orthodontics, Faculty of Dentistry, Federal University of Minas Gerais, Antoˆnio Carlos Avenue, 6627 Pampulha, Belo Horizonte, Minas Gerais 31270901, Brazil. Tel.: +55 31 3409 2470 Fax: +55 31 3409 2472 e-mail: [email protected] Submitted 13 June 2012; accepted 11 April 2013
tal caries experience in many countries, socioeconomic status is significantly associated with the problem. Dental caries prevalence differ between
289
Martins et al.
developed and developing countries as well as between different socioeconomic groups within countries (4, 5). In Brazil, data from the 2010 Oral Health Project reveal that 56.5% of 12-year-old Brazilian children have at least a permanent tooth with dental caries experience. This represents approximately 1.7 million children (6). In the city of Belo Horizonte, which is the capital of the state of Minas Gerais (southeastern Brazil), children aged 8 and 9 years have a high prevalence of toothache (45.9%) associated with social determinants and poor oral health status, and five-year-old children in the lowest social class have less access to dental treatment, suggesting a significant public health problem in the city (7, 8). In the multilevel approach, interactions between variables within a level and between levels can be modeled, avoiding the need for aggregating data to a higher level, which would result in the loss of statistical power and the risk of ecological fallacy (9). Despite its advantages, there are few studies that evaluate oral health and social factors in children, and fewer still that employ a multilevel approach. In a study conducted with children from one to 5 years of age to identify factors associated with oral health operating at the child, family, neighborhood, and state levels, the researchers found relatively few significant correlates of oral health at the state level whereas significant effects were found for variables measured at child, family, and neighborhood levels (10). In another study involving children from 1 to 5 years of age in Brazil, only individual variables exhibited significant associations with the presence of dental caries when all significant variables in the univariate analysis were incorporated into multivariate model. However, when the type of school was used alone, children enrolled in public schools had a significantly higher prevalence rate of dental caries than those enrolled in private schools (11). Considering the absence of prevalence studies in Brazil on dental caries in the mixed dentition phase, the aims of the present study were to assess caries experience among Brazilian children aged 8– 10 years and determine predisposing social factors of this disease, through a multilevel approach.
Methods Study design A cross-sectional survey was carried out with a representative sample (1204 male and female children
290
aged 8–10 years attending nineteen public and private elementary schools) in the city of Belo Horizonte, Minas Gerais, Brazil. Belo Horizonte has fluoride in the water supply and is divided into nine administration districts. Multistage sampling was employed for the selection of the children. The calculation of sample size was based on an a-value of 0.05, a detectable error of 3%, and an expected prevalence of 62.3%, resulting in 1002 children. Due to the lack of prevalence studies involving this age group, the prevalence value used for this calculation was based on a pilot study. This number was multiplied by 1.2 to compensate for the cluster effect (12) and an additional 20% was added to compensate for possible losses, leading to 1442 children (a total of 1436 statements of informed consent were distributed). To ensure representativity at the city level, the sample was stratified based on administrative district and type of institution. The percentage distribution of eight-to-ten-year-old schoolchildren in each administrative district was calculated from information provided by the local Board of Education. The distribution of participants was then determined by the proportion of this population in the respective school systems using data from samples (Table 1). The first-stage was comprised of randomly selected public and private elementary schools in each nine administrative district of Belo Horizonte. In the second-stage, classes were randomly chosen from the selected schools.
Data collection Two calibrated examiners [inter-examiner j = 0.78 (95% CI 0.73–0.83)] examined the children for the diagnosis of dental caries experience. Examinations were performed at school under natural light during daytime hours. The examiners used appropriate equipment to protect against individual cross-infection and all instruments (tongue depressor, dental mirror, and periodontal probe) were sterilized. No professional dental cleaning was performed prior to clinical exams. During the calibration process, 70 children (5% of the sample and not part of the main study population) were examined at a convenience school to determine interexaminer agreement and approximately 50 children were reexamined after 2 weeks to determine intraexaminer agreement. Excellent agreement was demonstrated [j = 0.93 (95% CI 0.91–0.96) and 0.95 (95% CI 0.93–0.97), respectively]. The parents’/caregiver’s signed informed consent and provided information on economic status. Socioeconomic status was determined using
Dental caries in Brazilian children: multilevel approach Table 1. Representative sample of children aged 8–10 years in Belo Horizonte, Brazil Administrative district of Belo Horizonte city Nordeste Public Private Noroeste Public Private Oeste Public Private Pampulha Public Private Venda Nova Public Private Barreiro Public Private Centro-Sul Public Private Leste Public Private Norte Public Private
Universe n = 97487 Number of students
Sample n = 1204 Proportion in universe (%)
Number of children
Proportion in sample (%)
9456 1462
9.7 1.5
116 19
9.6 1.6
9554 3510
9.8 3.6
118 43
9.8 3.6
7117 3217
7.3 3.3
89 40
7.4 3.3
5362 1462
5.5 1.5
66 20
5.5 1.7
10139 1657
10.4 1.7
124 20
10.3 1.7
10626 2047
10.9 2.1
131 25
10.8 2.1
6434 5557
6.6 5.7
82 67
6.8 5.6
8871 2437
9.1 2.5
110 30
9.1 2.4
7019 1560
7.2 1.6
85 19
7.1 1.6
household income, number of residents in the home, and parents’/caregivers’ schooling. Educational level was dichotomized into ≥8 years of schooling and 4); parents’/caregivers’ schooling (
Ó 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Factors associated with dental caries in Brazilian children: a multilevel approach
Milene T. Martins1, Fernanda Sardenberg1, Mauro H. Abreu2, Miriam P. Vale1, Saul M. Paiva1 and Isabela A . Pordeus1 1 Department of Pediatric Dentistry and Orthodontics, Faculty of Dentistry, Federal University of Minas Gerais, Belo Horizonte, Brazil, 2Department of Social and Preventive Dentistry, Faculty of Dentistry, Federal University of Minas Gerais, Belo Horizonte, Brazil
Martins MT, Sardenberg F, Abreu MH, Vale MP, Paiva SM, Pordeus IA. Factors associated with dental caries in Brazilian children: a multilevel approach. Community Dent Oral Epidemiol 2014; 42: 289–299. © 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: The aims of the present study were to assess caries experience among Brazilian children aged eight to 10 years and determine social factors of this disease, through a multilevel approach. Methods: This is a cross-sectional study in which a representative sample of 1204 eight-to-tenyear-old children were randomly selected from 19 public and private schools in the city of Belo Horizonte, Brazil. After formal authorization from the parents, the children were clinically examined at school by two calibrated examiners (Kappa=0.78). Individual socioeconomic status was determined using household income, number of residents in the home and parents’/caregivers’ schooling. The social vulnerability index and type of school were used as contextual variables. The outcome variables were severity score of dental caries in permanent teeth (DMFT index), dental caries experience in permanent teeth (DMFT = 0 or 6¼ 0), severity score of dental caries in primary teeth (dmft index) and dental caries experience in primary teeth (dmft = 0 or 6¼ 0). Results: The minority of children (37.2%) had dental caries experience. Chi-squared and Mann–Whitney tests showed that age, parents’/caregivers’ schooling, household income and type of school were associated with dental caries experience in permanent and primary teeth (P < 0.05). Mann–Whitney, Kruskal–Wallis tests and Spearman correlation coefficient showed that age, parents’/caregivers’ schooling, household income, and type of school were associated with severity score of dental caries in permanent and primary teeth (P < 0.05). In multilevel approach, the individual variables, age and household income demonstrated association with dental caries experience and with severity score of dental caries in primary teeth (P < 0.05). Type of school was also associated with dental caries experience and with severity score of dental caries in these teeth (P < 0.05).In the Poisson multilevel approach, the analysis of the final estimation of variance components of the null model revealed a significant difference among the 19 schools with regard to severity score of dental caries in permanent teeth too (P < 0.001). Younger children and attending private schools showed lower severity score of dental caries in permanent teeth. Conclusion: Dental caries experience and severity score of dental caries in primary teeth was influenced by age, household income, and the type of school The dental caries experience in permanent teeth was influenced by age, parents’/caregivers’ schooling, household income, and type of school, whereas the severity score of dental caries in permanent teeth was influenced by age and type of school.
Dental caries is highly prevalent in children (1, 2) and one of the most prevalent chronic diseases worldwide (3). Despite the reported decline of dendoi: 10.1111/cdoe.12087
Key words: dental caries; multilevel approach; social determinants MT Martins, Department of Pediatric Dentistry and Orthodontics, Faculty of Dentistry, Federal University of Minas Gerais, Antoˆnio Carlos Avenue, 6627 Pampulha, Belo Horizonte, Minas Gerais 31270901, Brazil. Tel.: +55 31 3409 2470 Fax: +55 31 3409 2472 e-mail: [email protected] Submitted 13 June 2012; accepted 11 April 2013
tal caries experience in many countries, socioeconomic status is significantly associated with the problem. Dental caries prevalence differ between
289
Martins et al.
developed and developing countries as well as between different socioeconomic groups within countries (4, 5). In Brazil, data from the 2010 Oral Health Project reveal that 56.5% of 12-year-old Brazilian children have at least a permanent tooth with dental caries experience. This represents approximately 1.7 million children (6). In the city of Belo Horizonte, which is the capital of the state of Minas Gerais (southeastern Brazil), children aged 8 and 9 years have a high prevalence of toothache (45.9%) associated with social determinants and poor oral health status, and five-year-old children in the lowest social class have less access to dental treatment, suggesting a significant public health problem in the city (7, 8). In the multilevel approach, interactions between variables within a level and between levels can be modeled, avoiding the need for aggregating data to a higher level, which would result in the loss of statistical power and the risk of ecological fallacy (9). Despite its advantages, there are few studies that evaluate oral health and social factors in children, and fewer still that employ a multilevel approach. In a study conducted with children from one to 5 years of age to identify factors associated with oral health operating at the child, family, neighborhood, and state levels, the researchers found relatively few significant correlates of oral health at the state level whereas significant effects were found for variables measured at child, family, and neighborhood levels (10). In another study involving children from 1 to 5 years of age in Brazil, only individual variables exhibited significant associations with the presence of dental caries when all significant variables in the univariate analysis were incorporated into multivariate model. However, when the type of school was used alone, children enrolled in public schools had a significantly higher prevalence rate of dental caries than those enrolled in private schools (11). Considering the absence of prevalence studies in Brazil on dental caries in the mixed dentition phase, the aims of the present study were to assess caries experience among Brazilian children aged 8– 10 years and determine predisposing social factors of this disease, through a multilevel approach.
Methods Study design A cross-sectional survey was carried out with a representative sample (1204 male and female children
290
aged 8–10 years attending nineteen public and private elementary schools) in the city of Belo Horizonte, Minas Gerais, Brazil. Belo Horizonte has fluoride in the water supply and is divided into nine administration districts. Multistage sampling was employed for the selection of the children. The calculation of sample size was based on an a-value of 0.05, a detectable error of 3%, and an expected prevalence of 62.3%, resulting in 1002 children. Due to the lack of prevalence studies involving this age group, the prevalence value used for this calculation was based on a pilot study. This number was multiplied by 1.2 to compensate for the cluster effect (12) and an additional 20% was added to compensate for possible losses, leading to 1442 children (a total of 1436 statements of informed consent were distributed). To ensure representativity at the city level, the sample was stratified based on administrative district and type of institution. The percentage distribution of eight-to-ten-year-old schoolchildren in each administrative district was calculated from information provided by the local Board of Education. The distribution of participants was then determined by the proportion of this population in the respective school systems using data from samples (Table 1). The first-stage was comprised of randomly selected public and private elementary schools in each nine administrative district of Belo Horizonte. In the second-stage, classes were randomly chosen from the selected schools.
Data collection Two calibrated examiners [inter-examiner j = 0.78 (95% CI 0.73–0.83)] examined the children for the diagnosis of dental caries experience. Examinations were performed at school under natural light during daytime hours. The examiners used appropriate equipment to protect against individual cross-infection and all instruments (tongue depressor, dental mirror, and periodontal probe) were sterilized. No professional dental cleaning was performed prior to clinical exams. During the calibration process, 70 children (5% of the sample and not part of the main study population) were examined at a convenience school to determine interexaminer agreement and approximately 50 children were reexamined after 2 weeks to determine intraexaminer agreement. Excellent agreement was demonstrated [j = 0.93 (95% CI 0.91–0.96) and 0.95 (95% CI 0.93–0.97), respectively]. The parents’/caregiver’s signed informed consent and provided information on economic status. Socioeconomic status was determined using
Dental caries in Brazilian children: multilevel approach Table 1. Representative sample of children aged 8–10 years in Belo Horizonte, Brazil Administrative district of Belo Horizonte city Nordeste Public Private Noroeste Public Private Oeste Public Private Pampulha Public Private Venda Nova Public Private Barreiro Public Private Centro-Sul Public Private Leste Public Private Norte Public Private
Universe n = 97487 Number of students
Sample n = 1204 Proportion in universe (%)
Number of children
Proportion in sample (%)
9456 1462
9.7 1.5
116 19
9.6 1.6
9554 3510
9.8 3.6
118 43
9.8 3.6
7117 3217
7.3 3.3
89 40
7.4 3.3
5362 1462
5.5 1.5
66 20
5.5 1.7
10139 1657
10.4 1.7
124 20
10.3 1.7
10626 2047
10.9 2.1
131 25
10.8 2.1
6434 5557
6.6 5.7
82 67
6.8 5.6
8871 2437
9.1 2.5
110 30
9.1 2.4
7019 1560
7.2 1.6
85 19
7.1 1.6
household income, number of residents in the home, and parents’/caregivers’ schooling. Educational level was dichotomized into ≥8 years of schooling and 4); parents’/caregivers’ schooling (
